Detection and Analysis of Digital Display Board Energy Consumption using IoT and Machine Learning Techniques

Nowadays Digital Display Boards (DDB) are used to post information in a variety of locations, including public spaces, hospitals, general stores, institutions, and colleges. Earlier, for displaying large data, the message needs to be changed for every instance. As of now digital displays are more preferred to static and attract the attention of viewers. The microcontroller present in the DDB write the information to the showing device. DDBs are connecting to the controller to continually scroll the message on the screen. The research article proposes the DDB system in office environments. The proposed system is fully designed and analyzed by IoT and Machine learning Techniques. The device will identify the DDB usage rate and reduce the wastage of DDB energy in unoccupied places and also forecasting future energy usage requirement. The proposed work applies a prediction system to detect and analyze the one Month energy consumption of DDB in the office environment and evaluates the existing model with the ARIMA algorithm for generating time-series based prediction models. To find out the precision of the proposed system, DDB along with sensor devices were installed in the office environment, which consist of current sensors, microcontrollers with cloud database connectivity. The set of data has been obtained from the database being utilized to evaluate and test the proposed models. according to the results of the prediction and analysis proposed DDB outperformed the ARIMA Model, with good accuracy. Based on the proposed method, the predicted accuracy value is 97.8% and R-squared for the model is 0.89. The Proposed DDB Energy Consumption system helps to monitor and detect the energy usage in office environments.